When an AI system makes the wrong diagnosis, misses a deterioration, recommends the wrong treatment or contributes to patient harm, who is ultimately responsible?

As AI moves from experimental technology into real clinical workflows, traditional lines of accountability are becoming increasingly blurred. Is liability held by the clinician who relied on the recommendation, the hospital that deployed the system, the developer who built the model, or the organisation that procured and governed it? And what happens when the system behaves in a way that even its developers cannot fully explain?

This panel brings together clinical, legal, technology and governance perspectives to explore what accountability should look like when AI is involved in patient care. We will examine where responsibility sits across the AI lifecycle, how human oversight should work in practice, and whether existing approaches to medical liability are fit for an AI-enabled healthcare system.

Most importantly, the discussion will ask a fundamental question: if everyone is responsible, is anyone actually accountable?